AI 精选动态
智能评分 68
LoopCoder-V2 发布
AI 推荐理由
版本提升明显，测试终身指标优异，技术实现细节有参考价值核心解读
该论文描述了新版 LoopCoder-V2 7B 规划模型的发布，重点介绍了其 shared Transformer 块优化与测试时间策略。同文中提到相比1-loop模型在Coding、软件工程、终端和工具使用场景上表现优于基准。加入了精简的代码训练和语言多样性数据。对比对象包括SOTA与上一代实现，展示了实质性改进。
全文
LoopCoder-V2 is here, a 7B PLT coding model that reuses shared Transformer blocks for one extra round of test-time reasoning. License: Apache 2.0.
👉 Try it now: https://t.co/L9QoZWGcip
✨ 2-loop sweet spot: stronger than the 1-loop baseline across coding, software engineering, terminal, and tool-use benchmarks
🛠️ Repo-level gains: biggest improvements show up on SWE-bench Verified and Multi-SWE
🌍 Code-first training: trained from scratch on 18T tokens with 100+ programming languages